Re: What Provosts Need to Mandate

From: Stevan Harnad <harnad_at_ecs.soton.ac.uk>
Date: Sat, 11 Dec 2004 13:13:21 +0000

>sh> it's quite obvious what needs to be done: Have a talk
>sh> with the university's provost... the one in the best position to
>sh> weigh the relative merits of the two conflicting university interests:
>sh> (1) to maximize university press journal revenue... and (2)
>sh> to maximize university research impact and income (as well as to invest
>sh> in perhaps eventually easing the university's serials budget burden)

On Sat, 11 Dec 2004, Ted Bergstrom wrote:

> It does seem like a good idea that somebody should be talking
> to university provosts about this. I will take a shot at...
> sending a message to... the UC provost.
> But who do you suggest should be bending the provostial ears at Chicago,
> Oxford, MIT, etc.?

There are researchers in the AmSci Forum from each of these universities
who are up-to-speed on the issue of maximising university research impact
through OA self-archiving. I hope they will speak to their provosts
(and pro-vice-chancellors) about resolving this conflict of interest
between each university's self-archiving needs and its university press's
self-archiving policy as soon possible.

>sh> the publisher's PDF is *unnecessary*: Just self-archive the author's
>sh> postprint and link it to the publisher's for-fee website for those who
>sh> may want the publisher's PDF; that's all that's needed -- and a more
>sh> than fair quid pro quo.)
>
> I agree that it is better to be able to self-archive one's own version
> of the final copy than not to be able to do so at all. But it is even
> better, both from the point of view of administrative convenience and
> of accuracy of the bibliographic record to be able to post the actual
> publishers' pdf file.

The best should not be allowed to get in the way of the better! At the moment,
we have neither:

Eighty percent of authors are not yet self-archiving at all, and most
universities do not yet have self-archiving policies. The explicit
rationale for Open Access (OA) is (1) that university research impact
is being needlessly lost worldwide because many would-be users at
many other universities worldwide are unable to access the journal in
which any given university's research output happens to be published,
so (2) each university needs to *supplement* what is currently accessible --
through subscription/license from the publisher to those universities
that can afford it -- with a supplementary OA version of each article,
self-archived by the author in his own university's OA archive.

Ninety-two percent of journals have now made the constructive decision to
give their authors the green light to self-archive their articles. This is
*not* the time to get needlessly greedy -- in the interest of the "best" over
the incomparably "better" situation that we would have if 100% of the annual
2.5 million research articles published in the world's 24,000 peer-reviewed
journals had a self-archived OA supplement for any would-be user whose university
cannot afford access to the journal in which it was published.

The author's own final, peer-reviewed draft fulfills the goal
of maximising access, usage and impact fully. The accuracy of the
bibliographic record continues to be ensured by the *journal version*,
whose PDF users can cite, even if they can only afford to access the
author's self-archived version.

And "administrative convenience" is hardly a defensible rationale for soliciting
further (needless) concessions from publishers. Lost research impact was a
justified and persuasive plaint; administrative inconvenience certainly is not!

Moreover, it is not the case that PDFs are easier or more convenient to process
than authors' postprints: I think you are thinking of retrospective "legacy"
archiving here, which is not the primary rationale for OA self-archiving: The
primary rationale -- and the place it needs to start -- is the self-archiving
of current and future research, not past research. Past research is
of course desirable and welcome, and will all come in due course, but
there is first a current hurdle to surmount, and that is not made easier
or more probable by handicapping it with the need to self-archive the
legacy literature too! Current and future OA provision will drive past
OA provision, in due time.

(And please, please don't forget that the purpose of OA self-archiving
is immediate access-provision and research impact-maximisation by providing
self-archived supplementary versions for those would-be users who cannot
afford access to the publisher's toll-access version! The purpose and
rationale for OA self-archiving is *not* long-term preservation! Of
course the supplements will be preserved too, but, until further notice,
the real preservation burden is on the publisher's primary version, and
on the libraries that license it, *not* on these secondary self-provided
supplements! So again, there is no need or justification for insisting
that they must be the publisher's PDFs.)

> As you are well aware, getting faculty to post their work is not as
> easy as you and I think it should be.

That is indeed true; in fact, that (and not publishers' unwillingness
to allow the self-archiving of their proprietary PDF) is *the* sole
substantive retardant on achieving immediate 100% OA today -- and has
been for a decade now!

But the solution is also equally apparent -- and we know it is the solution,
because when it has been implemented it has successfully generated 100% OA:
What is needed is a mandatory institutional/departmental self-archiving policy.
Where such a policy has been adopted, 100% OA has been reached quickly and
smoothly:

    http://archives.eprints.org/eprints.php?url=http%3A%2F%2Feprints.ecs.soton.ac.uk%2F
    http://software.eprints.org/handbook/departments.php
    http://www.eprints.org/signup/sign.php

Note that this policy specifically targets current and future research article
output, not (yet) legacy output (although that too is of course most welcome).

And researchers have already *told* us what they will and will not do: They will
not self-archive if they are not required to do so -- but if they *are* required
to do so, the vast majority say they will do so, and do so *willingly* (just as
they already comply willingly with the requirement to "publish or perish").

    Swan, A. & Brown, S.N. (2004a) JISC/OSI Journal Authors Survey Report.
    http://www.jisc.ac.uk/uploaded_documents/JISCOAreport1.pdf

    Swan, A. & Brown, S.N. (2004b) Authors and open access publishing.
    Learned Publishing 2004:17(3) 219-224.
    http://www.ingentaselect.com/rpsv/cw/alpsp/09531513/v17n3/s7/

Help in adopting such mandatory self-archiving policies will soon be
coming from the research funding agencies:

http://www.publications.parliament.uk/pa/cm200304/cmselect/cmsctech/399/39903.htm
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/3959.html
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/4115.html

But for the time being, universities are on their own in implementing
OA self-archiving policies; nor is there any reason at all that they
should wait for the new requirements from the research funders to become
official. University self-archiving is in any case naturally mandated,
monitored, measured and maintained at the local university/departmental
level, which is also where most of benefits of research impact (and the
costs of impact loss) are felt:

    "Re: Central versus institutional self-archiving"
    http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/3909.html

    Swan, A., Needham, P., Probets, S., Muir, A., O'Brien, A., Oppenheim,
    C., Hardy, R. & Rowland, F. (2004) Delivery, Management and Access
    Model for E-prints and Open Access Journals within Further and
    Higher Education.
    http://www.keyperspectives.co.uk/OpenAccessArchive/E-prints_delivery_model.pdf

> For those publishers who allow posting of the publishers' pdf, the
> administrative hassle is almost zero. A departmental administrator can
> simply ask the author for permission to put it on the website and put
> it up on the archive. It is a lot harder getting many authors to go to
> the trouble of assembling the last version of the paper, incorporating
> the final corrections, running it through Word or TeX and making a pdf
> file. To you or me this may not seem a big deal. To my colleagues? That
> is another story.

First, for current and forward-going literature this is far less of a hassle than
for retro: self-archiving the final, corrected, peer-reviewed draft can simply
become standard procedure, once a systematic and clear self-archiving policy is
formulated and adopted.

Besides, the relevant "hassles" to weigh against one another are the hassle of
generating a corrected final draft versus the hassle of having to haggle with the
publisher over PDF rights! In the interests of minimizing unnecessary hassle for
busy and still sluggish self-archivers, I strongly urge making self-archiving of
the corrected final draft mandatory, but the PDF only recommended, and optional.

> A final word on this matter. It would be helpful if those publishers
> who can not see their way to allow posting of recent pdf's would allow
> them to be posted with a lag of, say a year, like MIT Press. There can't
> be much gain in prohibiting posting of old articles. There ought to be
> some room for negotiation here.

This is again a legacy issue (and a low priority one, compared to the primary goal
of reaching 100% immediate OA). Minimize unnecessary negotiation; maximize
immediate self-archiving!

Fine to pursue the late PDFs too -- but not as an OA strategy, for OA
is defined as immediate access. Delayed or embargoed access is not OA
at all but just Shulenberger's 1998 NEAR proposal:

    "Shulenburger on open access: so NEAR and yet so far"
    http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/3277.html
    http://bmj.bmjjournals.com/cgi/eletters/328/7430/1#45401

    "Is Embarged Access Open Access?"
    http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/3687.html

    "AAAS's Response: Too Little, Too Late"
    http://www.sciencemag.org/cgi/eletters/291/5512/2318b

Stevan Harnad
Received on Sat Dec 11 2004 - 13:13:21 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:47:43 GMT